Overview 

Question Compatibility

1. Category
   
2. Line scale
   
3. Numeric
   
4. LMS
   

Questions/attributes with duplicate names in the same section of a test are excluded from the across-tests analysis.

Compatible Scenarios

When running analysis across tests, if the selected ANOVA type cannot be run (e.g. a 2-way option is selected but the data does not support it), the report will default to showing means only.

Generate Reports Across Tests

1. From the Dashboard, click Advanced search & analysis across tests.
   
   
2. Search for the tests you wish to run the report on and select them.
   
   
3. Click Reports.
   
   
4. If you would like to filter your data, in the top right-hand corner click Filters and use one, some, or all of the below filtering options:
   

   1. Sample sets. You can filter out sample sets for individual tests selected in the list.
      
      
      
   2. Samples. Samples from all tests are displayed in the tab. The test name they belong to will be displayed in the screen. Filter samples out as necessary.
      
      
   3. Reps. If any of the tests had reps defined in them, you will be able to filter reps out

   
   
5. From the Report type menu, select either Summary report or PowerPoint report.
   
   
   
6. Under the Options, do the following:  
   
1. Sample grouping options. The sample grouping option controls how samples across tests are matched for analysis. Sample grouping is applied at the test level and must produce a consistent data structure across all tests for the analysis to run successfully. If grouping results in multiple samples being assigned to the same group within a single test, their results will be averaged prior to analysis..
   
   It is important to select the correct grouping before generating the report.
   
   For report clarity you might want to ensure that the sample names are correct before running the reports. You can update sample names even when data is collected, but the test cannot be set to Complete. If you already set it to Complete and need to make changes in sample names, undo complete, update the samples, set the test back to Complete and run the analysis.
   
   
1. Group data with the same Sample name. Groups samples that share the same name across tests.
   
   
2. Group data by product. Groups samples by the product linked to samples.
   
   
3. Group data by Sample type. Groups samples by sample type.
   
   
4. Group data by custom Sample field. Groups samples by a custom field you specify.
   
   
5. Treat each sample as unique. No grouping. Each sample is treated as a distinct entry. 
   
   
2. Report format and details. Review the Report format and details options and make selections as needed.
   
   
3. Analysis options from defaults. The analysis options shown reflect your saved defaults. If the displayed ANOVA type or post hoc test does not match what you need for your across-tests analysis, click Change advanced analysis options to select the ANOVA and post hoc test that best align with your company's standards:
1. Analysis of variance - 1 way. The 1-way ANOVA compares samples not taking into account the panelist effect.
   
   
2. Analysis of variance - 2 way. The 2-way ANOVA compares samples taking into account the panelist/judge effect. Data from multiple tests is combined and analyzed together.
   
   
3. Analysis of variance - 2 way with interactions fixed - always use MSE. The 2-way ANOVA with interactions examines whether the Samples × Panelist interaction is significant. This is the default option when reps are present.
   
   
4. Analysis of variance - 2 way with interactions conditional - use MS interaction if interaction F-test is significant, otherwise use MSE. Similar to the fixed version, but switches the error term depending on whether the interaction is significant.
   
   
5. Multiple comparison test (post hoc) options. The post hoc test is applied after a significant ANOVA result to identify which samples differ from each other. Select from the following options:
   
1. Fisher's LSD test (significance p < 0.01, 0.05, 0.10, or 0.20)
   
2. Tukey's HSD test (significance p < 0.01, 0.05, 0.10, or 0.15)
   
3. Duncan's multiple range test (significance p < 0.01, 0.05, or 0.10)
   
4. Studentized Newman-Keuls (significance p < 0.01, 0.05, or 0.10)
   
5. Dunnett's test (significance p < 0.01, 0.05, or 0.10)
   
   Click Save to save the changes and return to the report generating screen.
   
   
7. Under the Questions, click Choose questions, review and select the questions you wish to include in the analysis.
   
   
8. Under the Export:
   
1. For Summary report, select the file format that best suits your reporting needs: .docx, .csv, or .dat.
   
   
2. For PowerPoint report, the only available option is .pptx. 
   
   Other report types will have different options to choose from.
   
   
9. Update the name for your report and click Create my report. Download and review the report.

When running analysis across tests, if the selected ANOVA type cannot be run (e.g. a 2-way option is selected but the data does not support it), the report will default to showing means only.

Changes made to the Advanced analysis options are shared across all report types in Compusense. This means that updating the ANOVA type or post hoc test will also be reflected in within-test reports until the settings are changed again.

Summary Report Details

1. Means table per sample group (with standard deviations if selected).
2. ANOVA p-values for the Samples effect (and Panelists/Judges where applicable).
3. Multiple comparison test results indicating which samples are significantly different.
4. Notes for any questions excluded from the analysis (not present in all tests, or duplicate question names).
   
   

PowerPoint Report Details

1. Means and standard deviations per sample group.
2. Multiple comparison test results (letters or values indicating significant differences).

Report Troubleshooting